name
string
code
string
asm
string
file
string
ncnn::BatchNorm_x86_avx2::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int BatchNorm_x86_avx2::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int dims = bottom_top_blob.dims; int elempack = bottom_top_blob.elempack; #if __SSE2__ #if __AVX__ if (elempack == 8) { if (dims == 1) { int w = bottom_top_blob.w; #pragma ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rsi, %rbx movq %rdi, %r14 movl 0x18(%rsi), %ecx movl 0x28(%rsi), %eax cmpl $0x4, %ecx je 0x4ae3a cmpl $0x8, %ecx jne 0x4ae99 cmpl $0x2, %eax je 0x4afa5 cmpl $0x1, %eax jne 0x4b019 movl 0x2c(%rbx), %ecx movq (%r14), %rdx xorl %esi, ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/batchnorm_x86_avx2.cpp
virtual thunk to ncnn::BatchNorm_x86_avx2::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int BatchNorm_x86_avx2::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int dims = bottom_top_blob.dims; int elempack = bottom_top_blob.elempack; #if __SSE2__ #if __AVX__ if (elempack == 8) { if (dims == 1) { int w = bottom_top_blob.w; #pragma ...
movq (%rdi), %rax addq -0x58(%rax), %rdi jmp 0x4adb0
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/batchnorm_x86_avx2.cpp
ncnn::Convolution_x86::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Convolution_x86::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const Mat& _weight_data = bottom_blobs[1]; Mat& top_blob = top_blobs[0]; const int _kernel_w = _weight_data.w; const int _kernel_h = _...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x168, %rsp # imm = 0x168 movq %rsi, %r12 movq %rdi, %r13 movq (%rsi), %r14 leaq 0x48(%r14), %rdi movq (%rdx), %rax movq %rax, 0x68(%rsp) movl 0x60(%r14), %ebx movl 0x74(%r14), %eax movl %eax, 0x1c(%rsp) movl 0x78(%r14), %eax movl %eax, ...
/tpoisonooo[P]ncnn/src/layer/x86/convolution_x86.cpp
ncnn::Convolution_x86_avx::Convolution_x86_avx()
Convolution_x86_avx::Convolution_x86_avx() { #if __SSE2__ support_packing = true; #if __AVX2__ support_weight_fp16_storage = true; #endif #endif // __SSE2__ activation = 0; convolution_dilation1 = 0; }
pushq %rbx movq %rdi, %rbx addq $0x180, %rdi # imm = 0x180 callq 0x4f6a4 leaq 0x19b295(%rip), %rax # 0x21b400 movq %rax, (%rbx) leaq 0x19b313(%rip), %rax # 0x21b488 movq %rax, 0x180(%rbx) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rbx) vmovups %xmm0, 0x1c(%rbx) vmovups %xmm0, 0x30(%rbx) vmovups %x...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::conv_im2col_sgemm_transform_kernel_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int)
static void conv_im2col_sgemm_transform_kernel_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_size) { const float* kernel = _kernel; // kernel memory packed 8 x 8 kernel_tm.create(8 * kernel_size, inch, outch / 8 + (outch % 8) / 4 + outch % 4); int nn_outch = 0; int remain...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xd8, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r14d movq %rsi, %r9 movq (%rdi), %rax movq %rax, 0x48(%rsp) leal (,%rbx,8), %esi pushq $0x8 popq %rcx movl %ebp, %eax cltd idivl %ecx movl %eax, %ecx movsbl %dl, %eax movb $0x4, %dl idivb %dl m...
/tpoisonooo[P]ncnn/src/layer/x86/convolution_sgemm.h
ncnn::InnerProduct_x86::forward_int8_x86(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int InnerProduct_x86::forward_int8_x86(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { const int num_input = weight_data_size / num_output; if (bottom_blob.dims == 2 && bottom_blob.w == num_input && bottom_blob.h * bottom_blob.elempack > 1) { // gemm Mat bottom_blob_unpack...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movq %rcx, 0x10(%rsp) movq %rdx, 0x18(%rsp) movq %rdi, %r15 movq (%rdi), %rax movq -0x18(%rax), %rcx movl 0xd8(%rdi,%rcx), %eax cltd idivl 0xd0(%rdi,%rcx) movl %eax, %r12d movq %rsi, %r13 cmpl $0x2, 0x28(%rsi) j...
/tpoisonooo[P]ncnn/src/layer/x86/innerproduct_x86.cpp
ncnn::InnerProduct_x86_avx2::InnerProduct_x86_avx2()
InnerProduct_x86_avx2::InnerProduct_x86_avx2() { #if __SSE2__ support_packing = true; #if __AVX2__ support_weight_fp16_storage = true; #endif #endif // __SSE2__ flatten = 0; activation = 0; }
movq (%rsi), %rax movq %rax, (%rdi) movq 0x8(%rsi), %rcx movq -0x18(%rax), %rax movq %rcx, (%rdi,%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x24(%rdi) vmovups %xmm0, 0x18(%rdi) vmovups %xmm0, 0x38(%rdi) vmovups %xmm0, 0x44(%rdi) vmovups %xmm0, 0x80(%rdi) vmovups %xmm0, 0x8c(%rdi) vmovups %xmm0, 0xc8(%rdi) vmovups ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx2::InnerProduct_x86_avx2()
InnerProduct_x86_avx2::InnerProduct_x86_avx2() { #if __SSE2__ support_packing = true; #if __AVX2__ support_weight_fp16_storage = true; #endif #endif // __SSE2__ flatten = 0; activation = 0; }
pushq %rbx movq %rdi, %rbx addq $0x138, %rdi # imm = 0x138 callq 0xac000 leaq 0x16bd49(%rip), %rax # 0x21c758 movq %rax, (%rbx) leaq 0x16bdbf(%rip), %rax # 0x21c7d8 movq %rax, 0x138(%rbx) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x18(%rbx) vmovups %xmm0, 0x24(%rbx) vmovups %xmm0, 0x38(%rbx) vmovups %x...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx2::create_pipeline(ncnn::Option const&)
int InnerProduct_x86_avx2::create_pipeline(const Option& opt) { // if (opt.use_packing_layout) { flatten = ncnn::create_layer(ncnn::LayerType::Flatten); ncnn::ParamDict pd; flatten->load_param(pd); flatten->create_pipeline(opt); } #if NCNN_INT8 if (opt.use_int8_in...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rsi, %rbx movq %rdi, %r14 pushq $0xe popq %rdi callq 0x3d2f3 movq %rax, 0x8(%r14) leaq 0x10(%rsp), %r15 movq %r15, %rdi callq 0x49b10 movq 0x8(%r14), %rdi movq (%rdi), %rax movq %r15, %rsi callq *0x10(%rax) movq 0x8(%r14), %rdi mov...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx2::create_pipeline_int8_x86(ncnn::Option const&)
int InnerProduct_x86_avx2::create_pipeline_int8_x86(const Option& opt) { activation = create_activation_layer(activation_type, activation_params, opt); const int num_input = weight_data_size / num_output; int out_elempack = 1; #if __SSE2__ if (opt.use_packing_layout) { out_elempack = num_o...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %r14 movq %rdi, %rbx movq (%rdi), %rax movq -0x18(%rax), %r13 movl 0xe0(%rdi,%r13), %ecx decl %ecx cmpl $0x5, %ecx ja 0xb0f75 leaq 0x11e963(%rip), %rax # 0x1cf654 movslq (%rax,%rcx,4), %rcx addq %rax, %rcx jmpq *%rcx pushq ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
virtual thunk to ncnn::InnerProduct_x86_avx2::create_pipeline(ncnn::Option const&)
int InnerProduct_x86_avx2::create_pipeline(const Option& opt) { // if (opt.use_packing_layout) { flatten = ncnn::create_layer(ncnn::LayerType::Flatten); ncnn::ParamDict pd; flatten->load_param(pd); flatten->create_pipeline(opt); } #if NCNN_INT8 if (opt.use_int8_in...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0xb0ab4 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx2::forward_int8_x86(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int InnerProduct_x86_avx2::forward_int8_x86(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { const int num_input = weight_data_size / num_output; if (bottom_blob.dims == 2 && bottom_blob.w == num_input && bottom_blob.h * bottom_blob.elempack > 1) { // gemm Mat bottom_blob_u...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movq %rcx, 0x10(%rsp) movq %rdx, 0x18(%rsp) movq %rdi, %r15 movq (%rdi), %rax movq -0x18(%rax), %rcx movl 0xd8(%rdi,%rcx), %eax cltd idivl 0xd0(%rdi,%rcx) movl %eax, %r12d movq %rsi, %r13 cmpl $0x2, 0x28(%rsi) j...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx2::forward_fp16(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int InnerProduct_x86_avx2::forward_fp16(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // flatten Mat bottom_blob_flattened = bottom_blob; if (bottom_blob.dims != 1) { Option opt_flatten = opt; opt_flatten.blob_allocator = opt.workspace_allocator; flatten->for...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x208, %rsp # imm = 0x208 movq %rcx, %r14 movq %rdx, %r15 movq %rsi, %r12 movq %rdi, %rbx leaq 0x178(%rsp), %rdi callq 0x462b0 cmpl $0x1, 0x28(%r12) je 0xbb1ba vmovups (%r14), %ymm0 vmovups 0x20(%r14), %ymm1 leaq 0xf0(%rsp), %rcx vmovups...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx2.cpp
ncnn::InnerProduct_x86_avx::InnerProduct_x86_avx()
InnerProduct_x86_avx::InnerProduct_x86_avx() { #if __SSE2__ support_packing = true; #if __AVX2__ support_weight_fp16_storage = true; #endif #endif // __SSE2__ flatten = 0; activation = 0; }
movq (%rsi), %rax movq %rax, (%rdi) movq 0x8(%rsi), %rcx movq -0x18(%rax), %rax movq %rcx, (%rdi,%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x24(%rdi) vmovups %xmm0, 0x18(%rdi) vmovups %xmm0, 0x38(%rdi) vmovups %xmm0, 0x44(%rdi) vmovups %xmm0, 0x80(%rdi) vmovups %xmm0, 0x8c(%rdi) vmovups %xmm0, 0xc8(%rdi) vmovups ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
ncnn::InnerProduct_x86_avx::InnerProduct_x86_avx()
InnerProduct_x86_avx::InnerProduct_x86_avx() { #if __SSE2__ support_packing = true; #if __AVX2__ support_weight_fp16_storage = true; #endif #endif // __SSE2__ flatten = 0; activation = 0; }
pushq %rbx movq %rdi, %rbx addq $0x138, %rdi # imm = 0x138 callq 0xac000 leaq 0x1603eb(%rip), %rax # 0x21c878 movq %rax, (%rbx) leaq 0x160461(%rip), %rax # 0x21c8f8 movq %rax, 0x138(%rbx) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x18(%rbx) vmovups %xmm0, 0x24(%rbx) vmovups %xmm0, 0x38(%rbx) vmovups %x...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
ncnn::InnerProduct_x86_avx::create_pipeline(ncnn::Option const&)
int InnerProduct_x86_avx::create_pipeline(const Option& opt) { // if (opt.use_packing_layout) { flatten = ncnn::create_layer(ncnn::LayerType::Flatten); ncnn::ParamDict pd; flatten->load_param(pd); flatten->create_pipeline(opt); } #if NCNN_INT8 if (opt.use_int8_inf...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %r15 movq %rdi, %rbx pushq $0xe popq %rdi callq 0x3d2f3 movq %rax, 0x8(%rbx) movq %rsp, %r14 movq %r14, %rdi callq 0x49b10 movq 0x8(%rbx), %rdi movq (%rdi), %rax movq %r14, %rsi callq *0x10(%rax) movq 0x8(%rbx), %rdi movq (%rd...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
ncnn::InnerProduct_x86_avx::create_pipeline_int8_x86(ncnn::Option const&)
int InnerProduct_x86_avx::create_pipeline_int8_x86(const Option& opt) { activation = create_activation_layer(activation_type, activation_params, opt); const int num_input = weight_data_size / num_output; int out_elempack = 1; #if __SSE2__ if (opt.use_packing_layout) { out_elempack = num_ou...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %r14 movq %rdi, %rbx movq (%rdi), %rax movq -0x18(%rax), %r13 movl 0xe0(%rdi,%r13), %ecx decl %ecx cmpl $0x5, %ecx ja 0xbc99d leaq 0x11310b(%rip), %rax # 0x1cf824 movslq (%rax,%rcx,4), %rcx addq %rax, %rcx jmpq *%rcx pushq ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
virtual thunk to ncnn::InnerProduct_x86_avx::create_pipeline(ncnn::Option const&)
int InnerProduct_x86_avx::create_pipeline(const Option& opt) { // if (opt.use_packing_layout) { flatten = ncnn::create_layer(ncnn::LayerType::Flatten); ncnn::ParamDict pd; flatten->load_param(pd); flatten->create_pipeline(opt); } #if NCNN_INT8 if (opt.use_int8_inf...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0xbc52a xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
ncnn::InnerProduct_x86_avx::forward_int8_x86(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int InnerProduct_x86_avx::forward_int8_x86(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { const int num_input = weight_data_size / num_output; if (bottom_blob.dims == 2 && bottom_blob.w == num_input && bottom_blob.h * bottom_blob.elempack > 1) { // gemm Mat bottom_blob_un...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movq %rcx, 0x10(%rsp) movq %rdx, 0x18(%rsp) movq %rdi, %r15 movq (%rdi), %rax movq -0x18(%rax), %rcx movl 0xd8(%rdi,%rcx), %eax cltd idivl 0xd0(%rdi,%rcx) movl %eax, %r12d movq %rsi, %r13 cmpl $0x2, 0x28(%rsi) j...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/innerproduct_x86_avx.cpp
ncnn::Input::Input()
Input::Input() { one_blob_only = true; support_inplace = true; support_vulkan = true; support_packing = true; support_bf16_storage = true; support_image_storage = true; }
pushq %rbx movq %rdi, %rbx callq 0x3d09a leaq 0x15477c(%rip), %rax # 0x21c990 movq %rax, (%rbx) movb $0x1, %al movb %al, 0xf(%rbx) movb %al, 0xc(%rbx) movl $0x1010101, 0x8(%rbx) # imm = 0x1010101 popq %rbx retq
/tpoisonooo[P]ncnn/src/layer/input.cpp
ncnn::Log::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Log::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; if (base == -1.f) { #pragma omp parallel for num_threads(opt.num_threads) for (int q = 0; q < chann...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x30(%rsi), %r13d movl 0x38(%rsi), %ebp movq %rsi, 0x8(%rsp) imull 0x2c(%rsi), %r13d movss 0xd0(%rdi), %xmm0 movss 0x106cda(%rip), %xmm1 # 0x1cf02c ucomiss %xmm0, %xmm1 jne 0xc83ca xorl %r15d, %r15d testl %r13d, %r...
/tpoisonooo[P]ncnn/src/layer/log.cpp
ncnn::LRN::load_param(ncnn::ParamDict const&)
int LRN::load_param(const ParamDict& pd) { region_type = pd.get(0, 0); local_size = pd.get(1, 5); alpha = pd.get(2, 1.f); beta = pd.get(3, 0.75f); bias = pd.get(4, 1.f); return 0; }
pushq %r14 pushq %rbx pushq %rax movq %rsi, %r14 movq %rdi, %rbx movq %rsi, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd0(%rbx) pushq $0x1 popq %rsi pushq $0x5 popq %rdx movq %r14, %rdi callq 0x49d20 movl %eax, 0xd4(%rbx) pushq $0x2 popq %rsi movss 0x1065b5(%rip), %xmm0 # 0x1cea88 movq %r14, %rdi...
/tpoisonooo[P]ncnn/src/layer/lrn.cpp
ncnn::LRN::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int LRN::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; size_t elemsize = bottom_top_blob.elemsize; int size = w * h; // squared values with local_size padding Mat square_blob; squa...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movl 0x2c(%rsi), %r11d movl 0x30(%rsi), %r10d movl 0x38(%rsi), %ecx movq %rsi, 0x38(%rsp) movq 0x10(%rsi), %r8 leaq 0xa0(%rsp), %rax andq $0x0, 0x40(%rax) movq %rdx, %r15 movq %rdi, %r14 xorps %xmm0, %xmm0 movap...
/tpoisonooo[P]ncnn/src/layer/lrn.cpp
ncnn::LRN_x86_avx2::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int LRN_x86_avx2::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; size_t elemsize = bottom_top_blob.elemsize; int size = w * h; // squared values with local_size padding Mat square_blob;...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1e8, %rsp # imm = 0x1E8 movl 0x2c(%rsi), %r10d movl 0x30(%rsi), %r8d movl 0x38(%rsi), %ecx movq %rsi, 0x38(%rsp) movq 0x10(%rsi), %r13 leaq 0xc0(%rsp), %rax andq $0x0, 0x40(%rax) movq %rdx, %r15 movq %rdi, %r14 vxorps %xmm0, %xmm0, %xm...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/lrn_x86_avx2.cpp
ncnn::LRN_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int LRN_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; size_t elemsize = bottom_top_blob.elemsize; int size = w * h; // squared values with local_size padding Mat square_blob; ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x178, %rsp # imm = 0x178 movl 0x2c(%rsi), %r10d movl 0x30(%rsi), %r8d movl 0x38(%rsi), %ecx movq %rsi, 0x38(%rsp) movq 0x10(%rsi), %r13 leaq 0xa0(%rsp), %rax andq $0x0, 0x40(%rax) movq %rdx, %r15 movq %rdi, %r14 vxorps %xmm0, %xmm0, %xm...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/lrn_x86_avx.cpp
ncnn::Pooling::load_param(ncnn::ParamDict const&)
int Pooling::load_param(const ParamDict& pd) { pooling_type = pd.get(0, 0); kernel_w = pd.get(1, 0); kernel_h = pd.get(11, kernel_w); stride_w = pd.get(2, 1); stride_h = pd.get(12, stride_w); pad_left = pd.get(3, 0); pad_right = pd.get(14, pad_left); pad_top = pd.get(13, pad_left); p...
pushq %rbp pushq %r14 pushq %rbx movq %rsi, %r14 movq %rdi, %rbx movq %rsi, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd0(%rbx) pushq $0x1 popq %rbp movq %r14, %rdi movl %ebp, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd4(%rbx) pushq $0xb popq %rsi movq %r14, %rdi movl %eax, %edx callq 0x49d20...
/tpoisonooo[P]ncnn/src/layer/pooling.cpp
ncnn::Pooling::make_padding(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
void Pooling::make_padding(const Mat& bottom_blob, Mat& bottom_blob_bordered, const Option& opt) const { int w = bottom_blob.w; int h = bottom_blob.h; bottom_blob_bordered = bottom_blob; float pad_value = 0.f; if (pooling_type == PoolMethod_MAX) { pad_value = bottom_blob.elemsize == 1 ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rcx, %r15 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r12 movl 0x2c(%rsi), %r13d movl 0x30(%rsi), %ebp movq %rdx, %rdi callq 0x3fd2a xorps %xmm0, %xmm0 cmpl $0x0, 0xd0(%r12) jne 0xcbe25 xorl %eax, %eax cmpq $0x1, 0x10(%r14) sete %a...
/tpoisonooo[P]ncnn/src/layer/pooling.cpp
virtual thunk to ncnn::Pooling_x86::create_pipeline(ncnn::Option const&)
int Pooling_x86::create_pipeline(const Option& /*opt*/) { if (adaptive_pooling) { support_packing = false; support_bf16_storage = false; support_fp16_storage = false; support_int8_storage = false; support_tensor_storage = false; support_weight_fp16_storage = fal...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0xcc0f8 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/src/layer/x86/pooling_x86.cpp
virtual thunk to ncnn::Pooling_x86::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Pooling_x86::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // max value in NxN window // avg value in NxN window if (adaptive_pooling) { return Pooling::forward(bottom_blob, top_blob, opt); } #if __SSE2__ int elempack = bottom_blob.elempack; int w = ...
movq (%rdi), %rax addq -0x48(%rax), %rdi jmp 0xcc14c nop
/tpoisonooo[P]ncnn/src/layer/x86/pooling_x86.cpp
ncnn::Pooling_x86_avx2::create_pipeline(ncnn::Option const&)
int Pooling_x86_avx2::create_pipeline(const Option& /*opt*/) { if (adaptive_pooling) { support_packing = false; support_bf16_storage = false; support_fp16_storage = false; support_int8_storage = false; support_tensor_storage = false; support_weight_fp16_storage ...
movq (%rdi), %rax movq -0x18(%rax), %rcx cmpl $0x0, 0x100(%rdi,%rcx) je 0xcce73 xorl %edx, %edx movb %dl, 0xb(%rdi,%rcx) movq -0x18(%rax), %rcx movb %dl, 0xc(%rdi,%rcx) movq -0x18(%rax), %rcx movb %dl, 0xd(%rdi,%rcx) movq -0x18(%rax), %rcx movb %dl, 0xe(%rdi,%rcx) movq -0x18(%rax), %rcx movb %dl, 0x10(%rdi,%rcx) movq -...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx2.cpp
virtual thunk to ncnn::Pooling_x86_avx2::create_pipeline(ncnn::Option const&)
int Pooling_x86_avx2::create_pipeline(const Option& /*opt*/) { if (adaptive_pooling) { support_packing = false; support_bf16_storage = false; support_fp16_storage = false; support_int8_storage = false; support_tensor_storage = false; support_weight_fp16_storage ...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0xcce34 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx2.cpp
virtual thunk to ncnn::Pooling_x86_avx2::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Pooling_x86_avx2::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // max value in NxN window // avg value in NxN window if (adaptive_pooling) { return Pooling::forward(bottom_blob, top_blob, opt); } #if __SSE2__ int elempack = bottom_blob.elempack; int...
movq (%rdi), %rax addq -0x48(%rax), %rdi jmp 0xcce88
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx2.cpp
virtual thunk to ncnn::Pooling_x86_avx::create_pipeline(ncnn::Option const&)
int Pooling_x86_avx::create_pipeline(const Option& /*opt*/) { if (adaptive_pooling) { support_packing = false; support_bf16_storage = false; support_fp16_storage = false; support_int8_storage = false; support_tensor_storage = false; support_weight_fp16_storage =...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0xceb34 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx.cpp
ncnn::Pooling_x86_avx::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Pooling_x86_avx::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // max value in NxN window // avg value in NxN window if (adaptive_pooling) { return Pooling::forward(bottom_blob, top_blob, opt); } #if __SSE2__ int elempack = bottom_blob.elempack; int ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x178, %rsp # imm = 0x178 movq %rcx, %r14 movq %rsi, %r12 movq %rdi, %r15 movq (%rdi), %rax movq -0x18(%rax), %rax addq %rax, %rdi cmpl $0x0, 0x100(%r15,%rax) je 0xcebd3 movq %r12, %rsi movq %r14, %rcx addq $0x178, %rsp # imm ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx.cpp
virtual thunk to ncnn::Pooling_x86_avx::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Pooling_x86_avx::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // max value in NxN window // avg value in NxN window if (adaptive_pooling) { return Pooling::forward(bottom_blob, top_blob, opt); } #if __SSE2__ int elempack = bottom_blob.elempack; int ...
movq (%rdi), %rax addq -0x48(%rax), %rdi jmp 0xceb88 nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/pooling_x86_avx.cpp
int ncnn::reduction<ncnn::reduction_op_add<float>, ncnn::reduction_op_add<float>, ncnn::post_process_identity<float>>(ncnn::Mat const&, ncnn::Mat&, float, bool, bool, bool, bool, float, int, ncnn::Option const&)
static int reduction(const Mat& a, Mat& b, float v0, bool reduce_w, bool reduce_h, bool reduce_c, bool post_process, float coeff, int keepdims, const Option& opt) { int ret; if (keepdims) ret = reduction_op_keepdims<Op, Op2>(a, b, v0, reduce_w, reduce_h, reduce_c, opt); else ret = reduction_...
pushq %rbp pushq %rbx pushq %rax movss %xmm0, 0x4(%rsp) movl %r9d, %ebp movq %rsi, %rbx movq 0x28(%rsp), %r9 movzbl %dl, %edx movzbl %cl, %ecx movzbl %r8b, %r8d cmpl $0x0, 0x20(%rsp) je 0xd8911 callq 0xd9ce2 jmp 0xd8916 callq 0xda6b4 testl %eax, %eax je 0xd891f pushq $-0x64 popq %rax jmp 0xd8952 testb %bpl, %bpl movss ...
/tpoisonooo[P]ncnn/src/layer/reduction.cpp
ncnn::ReLU_x86_avx2::forward_inplace_int8(ncnn::Mat&, ncnn::Option const&) const
int ReLU_x86_avx2::forward_inplace_int8(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int d = bottom_top_blob.d; int channels = bottom_top_blob.c; int size = w * h * d; int elempack = bottom_top_blob.elempack; #if __SSE2__ if (elempac...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %rbx movl 0x30(%rsi), %r12d imull 0x2c(%rsi), %r12d imull 0x34(%rsi), %r12d cmpl $0x8, 0x18(%rsi) movl 0x38(%rsi), %ebp movq (%rdi), %rax movq -0x18(%rax), %rax vmovss 0xd0(%rdi,%rax), %xmm0 vxorps %xmm1, %xmm1, %xmm1 jne 0xdd...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/relu_x86_avx2.cpp
ncnn::ReLU_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int ReLU_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int elembits = bottom_top_blob.elembits(); if (elembits == 8) return forward_inplace_int8(bottom_top_blob, opt); int w = bottom_top_blob.w; int h = bottom_top_blob.h; int d = bottom_top_blob.d; int chann...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rsi, %r12 movq %rdi, %r14 movl 0x18(%rsi), %ecx testl %ecx, %ecx movq %rsi, 0x8(%rsp) je 0xdd113 movl 0x10(%r12), %eax shll $0x3, %eax cltd idivl %ecx cmpl $0x8, %eax jne 0xdd12e movq %r14, %rdi movq %r12, %rsi callq 0xdd41c xorl %...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/relu_x86_avx.cpp
ncnn::ReLU_x86_avx::forward_inplace_int8(ncnn::Mat&, ncnn::Option const&) const
int ReLU_x86_avx::forward_inplace_int8(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int d = bottom_top_blob.d; int channels = bottom_top_blob.c; int size = w * h * d; int elempack = bottom_top_blob.elempack; #if __SSE2__ if (elempack...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %rbx movl 0x30(%rsi), %r12d imull 0x2c(%rsi), %r12d imull 0x34(%rsi), %r12d cmpl $0x8, 0x18(%rsi) movl 0x38(%rsi), %ebp movq (%rdi), %rax movq -0x18(%rax), %rax vmovss 0xd0(%rdi,%rax), %xmm0 vxorps %xmm1, %xmm1, %xmm1 jne 0xdd...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/relu_x86_avx.cpp
virtual thunk to ncnn::ReLU_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int ReLU_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int elembits = bottom_top_blob.elembits(); if (elembits == 8) return forward_inplace_int8(bottom_top_blob, opt); int w = bottom_top_blob.w; int h = bottom_top_blob.h; int d = bottom_top_blob.d; int chann...
pushq %rax movq (%rdi), %rax addq -0x58(%rax), %rdi callq 0xdd0c4 xorl %eax, %eax popq %rcx retq nopl (%rax)
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/relu_x86_avx.cpp
ncnn::Reshape::load_param(ncnn::ParamDict const&)
int Reshape::load_param(const ParamDict& pd) { w = pd.get(0, -233); h = pd.get(1, -233); d = pd.get(11, -233); c = pd.get(2, -233); permute = pd.get(3, 0); ndim = 4; if (d == -233) ndim = 3; if (c == -233) ndim = 2; if (h == -233) ndim = 1; if (w == -233)...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx movq %rsi, %r14 movq %rdi, %rbx xorl %r12d, %r12d movq %rsi, %rdi xorl %esi, %esi movl $0xffffff17, %edx # imm = 0xFFFFFF17 callq 0x49d20 movl %eax, 0xd0(%rbx) pushq $0x1 popq %rbp movq %r14, %rdi movl %ebp, %esi movl $0xffffff17, %edx # imm = 0xFFFFFF1...
/tpoisonooo[P]ncnn/src/layer/reshape.cpp
ncnn::Reshape::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Reshape::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { size_t elemsize = bottom_blob.elemsize; int total = bottom_blob.w * bottom_blob.h * bottom_blob.d * bottom_blob.c; int dims = bottom_blob.dims; // resolve out shape int outw = w; int outh = h; int outd ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x148, %rsp # imm = 0x148 movq %rcx, 0x30(%rsp) movq %rdx, %rbx movq 0x10(%rsi), %rax movq %rax, 0x48(%rsp) movl 0x2c(%rsi), %r9d movl 0x30(%rsi), %r8d movl %r8d, %eax imull %r9d, %eax movl 0x34(%rsi), %r14d movl 0x38(%rsi), %r11d movl %...
/tpoisonooo[P]ncnn/src/layer/reshape.cpp
ncnn::Reshape_x86::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Reshape_x86::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int elempack = bottom_blob.elempack; if (permute == 1) { // TODO implement permute on-the-fly Option opt_pack = opt; opt_pack.blob_allocator = opt.workspace_allocator; Mat bottom_blob...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x138, %rsp # imm = 0x138 movq %rcx, %r14 movq %rdx, %r13 movq %rsi, %r12 movq %rdi, %r15 movq (%rdi), %rax movq -0x18(%rax), %rcx cmpl $0x1, 0xe0(%rdi,%rcx) jne 0xde61d movdqu (%r14), %xmm0 movdqu 0x10(%r14), %xmm1 movups 0x20(%r14), %x...
/tpoisonooo[P]ncnn/src/layer/x86/reshape_x86.cpp
ncnn::Scale::load_model(ncnn::ModelBin const&)
int Scale::load_model(const ModelBin& mb) { if (scale_data_size == -233) return 0; scale_data = mb.load(scale_data_size, 1); if (scale_data.empty()) return -100; if (bias_term) { bias_data = mb.load(scale_data_size, 1); if (bias_data.empty()) return -100...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp movl 0xd0(%rdi), %edx xorl %ebp, %ebp cmpl $0xffffff17, %edx # imm = 0xFFFFFF17 je 0xe16ef movq %rsi, %r14 movq %rdi, %rbx movq (%rsi), %rax leaq 0x8(%rsp), %r12 pushq $0x1 popq %rcx movq %r12, %rdi callq *0x10(%rax) leaq 0xd8(%rbx), %r15 mov...
/tpoisonooo[P]ncnn/src/layer/scale.cpp
ncnn::Scale::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Scale::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { std::vector<Mat> bottom_top_blobs(2); bottom_top_blobs[0] = bottom_top_blob; bottom_top_blobs[1] = scale_data; return forward_inplace(bottom_top_blobs, opt); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x28, %rsp movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 leaq 0x10(%rsp), %r12 pushq $0x2 popq %rsi leaq 0xf(%rsp), %rdx movq %r12, %rdi callq 0xa0090 movq (%r12), %rdi movq %r15, %rsi callq 0x3fd2a leaq 0xd8(%r14), %rsi movq 0x10(%rsp), %rdi addq $0x48, %rdi callq 0x...
/tpoisonooo[P]ncnn/src/layer/scale.cpp
ncnn::Scale_x86::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %rbp movq (%rsi), %rbx movl 0x28(%rbx), %eax cmpl $0x4, 0x18(%rbx) jne 0xe1b10 cmpl $0x3, %eax je 0xe1c59 cmpl $0x2, %eax je 0xe1bd7 cmpl $0x1, %eax jne 0xe1e48 movl 0x2c(%rbx), %eax movq 0x48(%rbx), %rcx movq (%rbp), %rdx mov...
/tpoisonooo[P]ncnn/src/layer/x86/scale_x86.cpp
virtual thunk to ncnn::Scale_x86::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == ...
movq (%rdi), %rax addq -0x50(%rax), %rdi jmp 0xe1a80 nop
/tpoisonooo[P]ncnn/src/layer/x86/scale_x86.cpp
ncnn::Scale_x86_avx2::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86_avx2::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempac...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x68, %rsp movq %rdi, 0x8(%rsp) movq (%rsi), %rbx movl 0x18(%rbx), %ecx movl 0x28(%rbx), %eax cmpl $0x4, %ecx je 0xe1f59 cmpl $0x8, %ecx jne 0xe1fd4 cmpl $0x3, %eax je 0xe21f8 cmpl $0x2, %eax je 0xe20dc cmpl $0x1, %eax jne 0xe25c4 movl 0x2c(%rbx), ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/scale_x86_avx2.cpp
virtual thunk to ncnn::Scale_x86_avx2::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86_avx2::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempac...
movq (%rdi), %rax addq -0x50(%rax), %rdi jmp 0xe1eb0
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/scale_x86_avx2.cpp
ncnn::Scale_x86_avx::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86_avx::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x68, %rsp movq %rdi, 0x8(%rsp) movq (%rsi), %rbx movl 0x18(%rbx), %ecx movl 0x28(%rbx), %eax cmpl $0x4, %ecx je 0xe26d4 cmpl $0x8, %ecx jne 0xe274e cmpl $0x3, %eax je 0xe296b cmpl $0x2, %eax je 0xe2851 cmpl $0x1, %eax jne 0xe2d32 movl 0x2c(%rbx), ...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/scale_x86_avx.cpp
virtual thunk to ncnn::Scale_x86_avx::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Scale_x86_avx::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { Mat& bottom_top_blob = bottom_top_blobs[0]; const Mat& scale_blob = bottom_top_blobs[1]; int dims = bottom_top_blob.dims; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack...
movq (%rdi), %rax addq -0x50(%rax), %rdi jmp 0xe262c nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/scale_x86_avx.cpp
ncnn::Sigmoid::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Sigmoid::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #pragma omp parallel for num_threads(opt.num_threads) for (int q = 0; q < channels; q++) { float* p...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movl 0x30(%rsi), %r12d imull 0x2c(%rsi), %r12d xorl %r14d, %r14d testl %r12d, %r12d cmovlel %r14d, %r12d movq %rsi, 0x8(%rsp) movl 0x38(%rsi), %ebp testl %ebp, %ebp cmovlel %r14d, %ebp leaq 0x10(%rsp), %r15 cmpl %ebp, %r14d je 0xe2e01 mo...
/tpoisonooo[P]ncnn/src/layer/sigmoid.cpp
ncnn::Sigmoid_x86::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Sigmoid_x86::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq %rsi, %rbx movl 0x30(%rsi), %r15d movl 0x38(%rsi), %r12d imull 0x2c(%rsi), %r15d cmpl $0x4, 0x18(%rsi) jne 0xe2fde xorl %ebp, %ebp testl %r15d, %r15d cmovlel %ebp, %r15d testl %r12d, %r12d cmovlel %ebp, %r12d movq %rsp, %r14 cmpl %r...
/tpoisonooo[P]ncnn/src/layer/x86/sigmoid_x86.cpp
virtual thunk to ncnn::Sigmoid_x86::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Sigmoid_x86::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma ...
pushq %rax callq 0xe2e74 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/src/layer/x86/sigmoid_x86.cpp
void ncnn::qsort_descent_inplace<ncnn::BBoxRect>(std::vector<ncnn::BBoxRect, std::allocator<ncnn::BBoxRect>>&, std::vector<float, std::allocator<float>>&, int, int)
static void qsort_descent_inplace(std::vector<T>& datas, std::vector<float>& scores, int left, int right) { int i = left; int j = right; float p = scores[(left + right) / 2]; while (i <= j) { while (scores[i] > p) i++; while (scores[j] < p) j--; if ...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x20, %rsp movl %ecx, %ebx movl %edx, %ebp movq %rsi, %r14 movq %rdi, %r15 movq (%rsi), %rsi pushq $0x2 popq %r12 movl %ebp, %r8d leal (%r8,%rbx), %eax cltd idivl %r12d cltq movss (%rsi,%rax,4), %xmm0 movl %ebx, %r9d movl %ebp, %eax movl %r9d, %ecx movl %eax,...
/tpoisonooo[P]ncnn/src/layer/detectionoutput.cpp
ncnn::Interp::load_param(ncnn::ParamDict const&)
int Interp::load_param(const ParamDict& pd) { resize_type = pd.get(0, 0); height_scale = pd.get(1, 1.f); width_scale = pd.get(2, 1.f); output_height = pd.get(3, 0); output_width = pd.get(4, 0); dynamic_target_size = pd.get(5, 0); align_corner = pd.get(6, 0); if (resize_type < 0 || resiz...
pushq %r15 pushq %r14 pushq %rbx movq %rsi, %r15 movq %rdi, %r14 xorl %ebx, %ebx movq %rsi, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd0(%r14) pushq $0x1 popq %rsi movss 0x67e21(%rip), %xmm0 # 0x1cea88 movq %r15, %rdi callq 0x49d3a movss %xmm0, 0xd8(%r14) pushq $0x2 popq %rsi movq %r15, %rdi mo...
/tpoisonooo[P]ncnn/src/layer/interp.cpp
ncnn::Interp::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Interp::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const Mat& reference_blob = bottom_blobs[1]; Mat& top_blob = top_blobs[0]; int w = bottom_blob.w; int h = bottom_blob.h; int channels = bot...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x228, %rsp # imm = 0x228 movq %rdi, %r14 movq (%rsi), %rsi movq (%rdx), %r12 movl 0x2c(%rsi), %r10d movl 0x30(%rsi), %ebx movl 0x28(%rsi), %eax movq 0x10(%rsi), %r8 movslq 0x74(%rsi), %r11 pushq $-0x64 popq %r15 cmpl $0x2, %eax movq %rs...
/tpoisonooo[P]ncnn/src/layer/interp.cpp
ncnn::DeconvolutionDepthWise::load_param(ncnn::ParamDict const&)
int DeconvolutionDepthWise::load_param(const ParamDict& pd) { num_output = pd.get(0, 0); kernel_w = pd.get(1, 0); kernel_h = pd.get(11, kernel_w); dilation_w = pd.get(2, 1); dilation_h = pd.get(12, dilation_w); stride_w = pd.get(3, 1); stride_h = pd.get(13, stride_w); pad_left = pd.get(4...
pushq %rbp pushq %r14 pushq %rbx subq $0xa0, %rsp movq %rsi, %r14 movq %rdi, %rbx movq %rsi, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd0(%rbx) pushq $0x1 popq %rbp movq %r14, %rdi movl %ebp, %esi xorl %edx, %edx callq 0x49d20 movl %eax, 0xd4(%rbx) pushq $0xb popq %rsi movq %r14, %rdi movl %eax, %...
/tpoisonooo[P]ncnn/src/layer/deconvolutiondepthwise.cpp
ncnn::DeconvolutionDepthWise::load_model(ncnn::ModelBin const&)
int DeconvolutionDepthWise::load_model(const ModelBin& mb) { weight_data = mb.load(weight_data_size, 0); if (weight_data.empty()) return -100; if (bias_term) { bias_data = mb.load(num_output, 1); if (bias_data.empty()) return -100; } return 0; }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp movq %rsi, %r14 movq %rdi, %rbx movl 0x110(%rdi), %edx movq (%rsi), %rax leaq 0x8(%rsp), %r12 movq %r12, %rdi xorl %ecx, %ecx callq *0x10(%rax) leaq 0x168(%rbx), %r15 movq %r15, %rdi movq %r12, %rsi callq 0x3fd2a leaq 0x8(%rsp), %rdi callq 0x3fd8e ...
/tpoisonooo[P]ncnn/src/layer/deconvolutiondepthwise.cpp
ncnn::DeconvolutionDepthWise::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int DeconvolutionDepthWise::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { // deconvolv with NxN kernel // value = value + bias int w = bottom_blob.w; int h = bottom_blob.h; int channels = bottom_blob.c; size_t elemsize = bottom_blob.elemsize; if (channels % grou...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1a8, %rsp # imm = 0x1A8 movq %rdx, %r8 movl 0x38(%rsi), %r10d movl 0x114(%rdi), %r9d movl %r10d, %eax cltd idivl %r9d pushq $-0x64 popq %rbp testl %edx, %edx jne 0x168db1 movq %rcx, 0x58(%rsp) movq %r8, 0x50(%rsp) movl 0xd0(%rdi), %ecx...
/tpoisonooo[P]ncnn/src/layer/deconvolutiondepthwise.cpp
ncnn::DeconvolutionDepthWise::cut_padding(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
void DeconvolutionDepthWise::cut_padding(const Mat& top_blob_bordered, Mat& top_blob, const Option& opt) const { if (pad_left > 0 || pad_right > 0 || pad_top > 0 || pad_bottom > 0) { copy_cut_border(top_blob_bordered, top_blob, pad_top, pad_bottom, pad_left, pad_right, opt); } else if (output_w ...
pushq %rbp pushq %r14 pushq %rbx subq $0x10, %rsp movq %rdx, %rbx movl 0xec(%rdi), %r8d movl 0xf0(%rdi), %r9d testl %r8d, %r8d setg %al testl %r9d, %r9d setg %dl orb %al, %dl jne 0x168e43 movl 0xf4(%rdi), %eax testl %eax, %eax jg 0x168e43 movl 0xf8(%rdi), %edx testl %edx, %edx jle 0x168e69 movl 0xf4(%rdi), %edx movl 0x...
/tpoisonooo[P]ncnn/src/layer/deconvolutiondepthwise.cpp
ncnn::ShuffleChannel::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int ShuffleChannel::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int w = bottom_blob.w; int h = bottom_blob.h; int channels = bottom_blob.c; size_t elemsize = bottom_blob.elemsize; if (channels % group != 0) { // reject invalid group return -100; ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xc8, %rsp movq %rdx, %r15 movq %rsi, %r12 movl 0x38(%rsi), %r8d movl 0xd0(%rdi), %esi movl %r8d, %eax cltd idivl %esi pushq $-0x64 popq %rbp testl %edx, %edx je 0x169008 movl %ebp, %eax addq $0xc8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq...
/tpoisonooo[P]ncnn/src/layer/shufflechannel.cpp
ncnn::InstanceNorm::load_model(ncnn::ModelBin const&)
int InstanceNorm::load_model(const ModelBin& mb) { if (affine == 0) return 0; gamma_data = mb.load(channels, 1); if (gamma_data.empty()) return -100; beta_data = mb.load(channels, 1); if (beta_data.empty()) return -100; return 0; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp xorl %ebx, %ebx cmpl $0x0, 0xd8(%rdi) je 0x1692ef movq %rsi, %r15 movq %rdi, %r14 movl 0xd0(%rdi), %edx movq (%rsi), %rax movq %rsp, %r13 pushq $0x1 popq %rcx movq %r13, %rdi callq *0x10(%rax) leaq 0xe0(%r14), %r12 movq %r12, %rdi movq %...
/tpoisonooo[P]ncnn/src/layer/instancenorm.cpp
ncnn::InstanceNorm::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int InstanceNorm::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { // x = (x - mean) / (sqrt(var + eps)) * gamma + beta int w = bottom_top_blob.w; int h = bottom_top_blob.h; int c = bottom_top_blob.c; int size = w * h; #pragma omp parallel for num_threads(opt.num_threads) f...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x30(%rsi), %r13d movl 0x38(%rsi), %ebp movq %rsi, 0x8(%rsp) imull 0x2c(%rsi), %r13d cvtsi2ss %r13d, %xmm0 xorl %r15d, %r15d testl %r13d, %r13d cmovlel %r15d, %r13d testl %ebp, %ebp cmovlel %r15d, %ebp movss 0x65733(...
/tpoisonooo[P]ncnn/src/layer/instancenorm.cpp
ncnn::Clip::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #pragma omp parallel for num_threads(opt.num_threads) for (int q = 0; q < channels; q++) { float* ptr ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x30(%rsi), %r13d movl 0x38(%rsi), %ebp movq %rsi, 0x8(%rsp) imull 0x2c(%rsi), %r13d xorl %r15d, %r15d testl %r13d, %r13d cmovlel %r15d, %r13d testl %ebp, %ebp cmovlel %r15d, %ebp leaq 0x10(%rsp), %r12 cmpl %ebp, %r1...
/tpoisonooo[P]ncnn/src/layer/clip.cpp
ncnn::Clip_x86::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma omp...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x30(%rsi), %r12d movl 0x38(%rsi), %r13d imull 0x2c(%rsi), %r12d movq %rsi, 0x8(%rsp) cmpl $0x4, 0x18(%rsi) jne 0x1696c6 xorl %ebp, %ebp testl %r12d, %r12d cmovlel %ebp, %r12d testl %r13d, %r13d cmovlel %ebp, %r13d l...
/tpoisonooo[P]ncnn/src/layer/x86/clip_x86.cpp
virtual thunk to ncnn::Clip_x86::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma omp...
pushq %rax movq (%rdi), %rax addq -0x58(%rax), %rdi callq 0x169620 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/src/layer/x86/clip_x86.cpp
ncnn::Clip_x86_avx2::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86_avx2::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragm...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x18(%rsi), %eax movl 0x30(%rsi), %r12d movl 0x38(%rsi), %r13d movq %rsi, 0x8(%rsp) imull 0x2c(%rsi), %r12d xorl %ebp, %ebp cmpl $0x8, %eax je 0x16985d cmpl $0x4, %eax jne 0x1698d3 testl %r12d, %r12d cmovlel %ebp, %r...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/clip_x86_avx2.cpp
virtual thunk to ncnn::Clip_x86_avx2::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86_avx2::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragm...
pushq %rax movq (%rdi), %rax addq -0x58(%rax), %rdi callq 0x1697b4 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/clip_x86_avx2.cpp
ncnn::Clip_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x18(%rsi), %eax movl 0x30(%rsi), %r12d movl 0x38(%rsi), %r13d movq %rsi, 0x8(%rsp) imull 0x2c(%rsi), %r12d xorl %ebp, %ebp cmpl $0x8, %eax je 0x169a69 cmpl $0x4, %eax jne 0x169adf testl %r12d, %r12d cmovlel %ebp, %r...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/clip_x86_avx.cpp
virtual thunk to ncnn::Clip_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Clip_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragma...
pushq %rax movq (%rdi), %rax addq -0x58(%rax), %rdi callq 0x1699c0 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/clip_x86_avx.cpp
ncnn::Reorg::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Reorg::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int w = bottom_blob.w; int h = bottom_blob.h; int channels = bottom_blob.c; size_t elemsize = bottom_blob.elemsize; int outw = w / stride; int outh = h / stride; int outc = channels * stride * stride; ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xc8, %rsp movq %rdx, %r15 movl 0x2c(%rsi), %eax movl 0x30(%rsi), %r9d movl 0x38(%rsi), %r10d movq %rsi, 0x30(%rsp) movq 0x10(%rsi), %r8 movq %rdi, 0x10(%rsp) movl 0xd0(%rdi), %edi cltd idivl %edi movl %eax, %ebp movl %r9d, %eax cltd idivl %edi mov...
/tpoisonooo[P]ncnn/src/layer/reorg.cpp
ncnn::YoloDetectionOutput::forward_inplace(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int YoloDetectionOutput::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const { // gather all box std::vector<BBoxRect> all_bbox_rects; std::vector<float> all_bbox_scores; for (size_t b = 0; b < bottom_top_blobs.size(); b++) { Mat& bottom_top_blob = bottom_top_blobs[...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x208, %rsp # imm = 0x208 movq %rdx, 0xb0(%rsp) movq %rdi, %rbp xorps %xmm0, %xmm0 leaq 0x40(%rsp), %rax andq $0x0, 0x10(%rax) leaq 0x90(%rsp), %rcx andq $0x0, 0x10(%rcx) movaps %xmm0, (%rax) movaps %xmm0, (%rcx) pushq $0x48 popq %rcx le...
/tpoisonooo[P]ncnn/src/layer/yolodetectionoutput.cpp
void ncnn::qsort_descent_inplace<ncnn::BBoxRect>(std::vector<ncnn::BBoxRect, std::allocator<ncnn::BBoxRect>>&, std::vector<float, std::allocator<float>>&, int, int)
static void qsort_descent_inplace(std::vector<T>& datas, std::vector<float>& scores, int left, int right) { int i = left; int j = right; float p = scores[(left + right) / 2]; while (i <= j) { while (scores[i] > p) i++; while (scores[j] < p) j--; if ...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x20, %rsp movl %ecx, %ebx movl %edx, %ebp movq %rsi, %r14 movq %rdi, %r15 movq (%rsi), %rsi pushq $0x2 popq %r12 movl %ebp, %r8d leal (%r8,%rbx), %eax cltd idivl %r12d cltq movss (%rsi,%rax,4), %xmm0 movl %ebx, %r9d movl %ebp, %eax movl %r9d, %ecx movl %eax,...
/tpoisonooo[P]ncnn/src/layer/yolodetectionoutput.cpp
ncnn::Quantize::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Quantize::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int dims = bottom_blob.dims; if (dims == 1) { int w = bottom_blob.w; top_blob.create(w, (size_t)1u, opt.blob_allocator); if (top_blob.empty()) return -100; const float* ptr ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x88, %rsp movq %rdx, %r15 movq %rsi, 0x18(%rsp) movl 0x28(%rsi), %eax cmpl $0x3, %eax movq %rdi, 0x38(%rsp) je 0x16ae2a movq %rdi, %r12 cmpl $0x2, %eax je 0x16ad09 movq $0x0, 0x10(%rsp) cmpl $0x1, %eax jne 0x16afba movq 0x18(%rsp), %rax movl 0x2c(...
/tpoisonooo[P]ncnn/src/layer/quantize.cpp
ncnn::Dequantize::load_model(ncnn::ModelBin const&)
int Dequantize::load_model(const ModelBin& mb) { scale_data = mb.load(scale_data_size, 1); if (scale_data.empty()) return -100; if (bias_data_size) { bias_data = mb.load(bias_data_size, 1); if (bias_data.empty()) return -100; } return 0; }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp movq %rsi, %r14 movq %rdi, %rbx movl 0xd0(%rdi), %edx movq (%rsi), %rax leaq 0x8(%rsp), %r12 pushq $0x1 popq %rcx movq %r12, %rdi callq *0x10(%rax) leaq 0xd8(%rbx), %r15 movq %r15, %rdi movq %r12, %rsi callq 0x3fd2a leaq 0x8(%rsp), %rdi callq 0x3fd...
/tpoisonooo[P]ncnn/src/layer/dequantize.cpp
ncnn::Yolov3DetectionOutput::load_param(ncnn::ParamDict const&)
int Yolov3DetectionOutput::load_param(const ParamDict& pd) { num_class = pd.get(0, 20); num_box = pd.get(1, 5); confidence_threshold = pd.get(2, 0.01f); nms_threshold = pd.get(3, 0.45f); biases = pd.get(4, Mat()); mask = pd.get(5, Mat()); anchors_scale = pd.get(6, Mat()); return 0; }
pushq %r15 pushq %r14 pushq %rbx subq $0xa0, %rsp movq %rsi, %r14 movq %rdi, %rbx pushq $0x14 popq %rdx movq %rsi, %rdi xorl %esi, %esi callq 0x49d20 movl %eax, 0xd0(%rbx) pushq $0x1 popq %rsi pushq $0x5 popq %rdx movq %r14, %rdi callq 0x49d20 movl %eax, 0xd4(%rbx) pushq $0x2 popq %rsi movss 0x5b9c2(%rip), %xmm0 # 0...
/tpoisonooo[P]ncnn/src/layer/yolov3detectionoutput.cpp
virtual thunk to ncnn::Yolov3DetectionOutput_x86_avx2::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Yolov3DetectionOutput_x86_avx2::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { // gather all box std::vector<BBoxRect> all_bbox_rects; for (size_t b = 0; b < bottom_blobs.size(); b++) { std::vector<std::vector<BBoxRect> > all_box_bbox_r...
movq (%rdi), %rax addq -0x40(%rax), %rdi jmp 0x176828 nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/yolov3detectionoutput_x86_avx2.cpp
virtual thunk to ncnn::Yolov3DetectionOutput_x86_avx::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Yolov3DetectionOutput_x86_avx::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { // gather all box std::vector<BBoxRect> all_bbox_rects; for (size_t b = 0; b < bottom_blobs.size(); b++) { std::vector<std::vector<BBoxRect> > all_box_bbox_re...
movq (%rdi), %rax addq -0x40(%rax), %rdi jmp 0x1771f8 nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/yolov3detectionoutput_x86_avx.cpp
ncnn::ROIAlign_x86::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int ROIAlign_x86::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const int width = bottom_blob.w; const int height = bottom_blob.h; const size_t elemsize = bottom_blob.elemsize; const int channels = bott...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x148, %rsp # imm = 0x148 movq %rdi, %r14 movq (%rsi), %rax movl 0x2c(%rax), %ebp movl 0x30(%rax), %r12d movq 0x10(%rax), %r8 movq %rax, 0x60(%rsp) movl 0x38(%rax), %r10d movq (%rdx), %r15 movq (%rdi), %rax movq -0x18(%rax), %rax movl 0x...
/tpoisonooo[P]ncnn/src/layer/x86/roialign_x86.cpp
virtual thunk to ncnn::ROIAlign_x86_avx::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int ROIAlign_x86_avx::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const int width = bottom_blob.w; const int height = bottom_blob.h; const size_t elemsize = bottom_blob.elemsize; const int channels = ...
movq (%rdi), %rax addq -0x40(%rax), %rdi jmp 0x17a040
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/roialign_x86_avx.cpp
ncnn::Packing_x86_avx::forward_int8(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Packing_x86_avx::forward_int8(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { if (use_padding) { return Packing::forward(bottom_blob, top_blob, opt); } size_t elemsize = bottom_blob.elemsize; int elempack = bottom_blob.elempack; if (elempack == out_elempack) {...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rdx, %r11 movq %rsi, %r14 movq %rdi, %r13 movq (%rdi), %rax movq -0x18(%rax), %rax addq %rax, %rdi cmpl $0x0, 0xd4(%r13,%rax) je 0x180645 movq %r14, %rsi movq %r11, %rdx addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq...
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/packing_x86_avx.cpp
ncnn::Requantize::load_param(ncnn::ParamDict const&)
int Requantize::load_param(const ParamDict& pd) { // scale_in = pd.get(0, 1.f); // bottom_blob_scale * weight_scale // scale_out = pd.get(1, 1.f); // top_blob_scale // bias_term = pd.get(2, 0); // bias_data_size = pd.get(3, 0); scale_in_data_size = pd.get(0, 1); scale_out_data_...
pushq %rbp pushq %r14 pushq %rbx subq $0xa0, %rsp movq %rsi, %r14 movq %rdi, %rbx pushq $0x1 popq %rbp movq %rsi, %rdi xorl %esi, %esi movl %ebp, %edx callq 0x49d20 movl %eax, 0xd0(%rbx) movq %r14, %rdi movl %ebp, %esi movl %ebp, %edx callq 0x49d20 movl %eax, 0xd4(%rbx) pushq $0x2 popq %rsi movq %r14, %rdi xorl %edx, %...
/tpoisonooo[P]ncnn/src/layer/requantize.cpp
ncnn::Requantize::load_model(ncnn::ModelBin const&)
int Requantize::load_model(const ModelBin& mb) { scale_in_data = mb.load(scale_in_data_size, 1); if (scale_in_data.empty()) return -100; scale_out_data = mb.load(scale_out_data_size, 1); if (scale_out_data.empty()) return -100; if (bias_data_size) { bias_data = mb.load(...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp movq %rsi, %r14 movq %rdi, %rbx movl 0xd0(%rdi), %edx movq (%rsi), %rax leaq 0x8(%rsp), %r12 pushq $0x1 popq %rcx movq %r12, %rdi callq *0x10(%rax) leaq 0x128(%rbx), %r15 movq %r15, %rdi movq %r12, %rsi callq 0x3fd2a leaq 0x8(%rsp), %rdi callq 0x3f...
/tpoisonooo[P]ncnn/src/layer/requantize.cpp
ncnn::Requantize::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Requantize::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int dims = bottom_blob.dims; if (dims == 1) { int w = bottom_blob.w; top_blob.create(w, (size_t)1u, opt.blob_allocator); if (top_blob.empty()) return -100; const int* intp...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movq %rdx, %r12 movq %rdi, %rbx movq %rsi, 0x10(%rsp) movl 0x28(%rsi), %edx pushq $-0x64 popq %r14 cmpl $0x3, %edx movq %r12, 0x20(%rsp) je 0x18155f cmpl $0x2, %edx je 0x1813ed xorl %eax, %eax cmpl $0x1, %edx jne 0x181e63 movq 0x10(%rsp)...
/tpoisonooo[P]ncnn/src/layer/requantize.cpp
ncnn::Cast::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Cast::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { if (type_from == type_to) { top_blob = bottom_blob; return 0; } int w = bottom_blob.w; int h = bottom_blob.h; int d = bottom_blob.d; int channels = bottom_blob.c; int dims = bottom_blob.d...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movq %rdx, %r14 movq %rsi, %rbx movl 0xd4(%rdi), %edx cmpl %edx, 0xd0(%rdi) jne 0x1b6185 movq %r14, %rdi movq %rbx, %rsi callq 0x3fd2a movq $0x0, 0x10(%rsp) jmp 0x1b6624 movq %rcx, %rax movq %rdi, %r12 movl 0x28(%rbx), %ecx movslq 0x18(%...
/tpoisonooo[P]ncnn/src/layer/cast.cpp
ncnn::HardSigmoid_x86::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int HardSigmoid_x86::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pra...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdi, %r14 movl 0x30(%rsi), %r12d movl 0x38(%rsi), %r13d imull 0x2c(%rsi), %r12d xorl %r15d, %r15d testl %r12d, %r12d cmovlel %r15d, %r12d testl %r13d, %r13d cmovlel %r15d, %r13d movq %rsi, 0x8(%rsp) cmpl $0x4, 0x18(%rsi) jne 0x1b70...
/tpoisonooo[P]ncnn/src/layer/x86/hardsigmoid_x86.cpp
ncnn::Swish::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Swish::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #pragma omp parallel for num_threads(opt.num_threads) for (int q = 0; q < channels; q++) { float* ptr...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x68, %rsp movl 0x30(%rsi), %r12d imull 0x2c(%rsi), %r12d xorl %r14d, %r14d testl %r12d, %r12d cmovlel %r14d, %r12d movq %rsi, 0x8(%rsp) movl 0x38(%rsi), %ebp testl %ebp, %ebp cmovlel %r14d, %ebp leaq 0x20(%rsp), %r15 cmpl %ebp, %r14d je 0x1ba7b3 m...
/tpoisonooo[P]ncnn/src/layer/swish.cpp
virtual thunk to ncnn::Swish_x86_avx::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int Swish_x86_avx::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { int w = bottom_top_blob.w; int h = bottom_top_blob.h; int channels = bottom_top_blob.c; int size = w * h; #if __SSE2__ int elempack = bottom_top_blob.elempack; #if __AVX__ if (elempack == 8) { #pragm...
pushq %rax callq 0x1bb164 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/build_O2/src/layer/x86/swish_x86_avx.cpp
ncnn::LayerNorm::forward_inplace(ncnn::Mat&, ncnn::Option const&) const
int LayerNorm::forward_inplace(Mat& bottom_top_blob, const Option& opt) const { // x = (x - mean) / sqrt(var + eps) * gamma + beta int dims = bottom_top_blob.dims; if (dims == 2) { int w = bottom_top_blob.w; int h = bottom_top_blob.h; // assert affine_size == w #pragma...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rsi, %r14 movq %rdi, %rbx movl 0x28(%rsi), %eax cmpl $0x3, %eax je 0x1bc3de cmpl $0x2, %eax jne 0x1bc521 movslq 0x2c(%r14), %rax movl 0x30(%r14), %ecx cvtsi2ss %eax, %xmm1 xorl %edx, %edx testl %eax, %eax movl $0x0, %esi cmovgl %ea...
/tpoisonooo[P]ncnn/src/layer/layernorm.cpp
ncnn::MultiHeadAttention::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int MultiHeadAttention::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& q_blob = bottom_blobs[0]; const Mat& k_blob = bottom_blobs.size() == 1 ? q_blob : bottom_blobs[1]; const Mat& v_blob = bottom_blobs.size() == 1 ? q_blob : bottom_blobs[2];...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x2a8, %rsp # imm = 0x2A8 movq %rdx, %r8 movq (%rsi), %rbp movq 0x8(%rsi), %rax subq %rbp, %rax xorl %edx, %edx cmpq $0x48, %rax pushq $0x48 popq %rax cmoveq %rdx, %rax movq %rax, 0xa8(%rsp) movl $0x90, %r14d cmoveq %rdx, %r14 movl 0x30(...
/tpoisonooo[P]ncnn/src/layer/multiheadattention.cpp
ncnn::Convolution1D::load_model(ncnn::ModelBin const&)
int Convolution1D::load_model(const ModelBin& mb) { if (dynamic_weight) return 0; weight_data = mb.load(weight_data_size, 0); if (weight_data.empty()) return -100; if (bias_term) { bias_data = mb.load(num_output, 1); if (bias_data.empty()) return -100; ...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp xorl %ebp, %ebp cmpl $0x0, 0x140(%rdi) je 0x1be98e movl %ebp, %eax addq $0x50, %rsp popq %rbx popq %r12 popq %r14 popq %r15 popq %rbp retq movq %rsi, %r14 movq %rdi, %rbx movl 0xf0(%rdi), %edx movq (%rsi), %rax leaq 0x8(%rsp), %r12 movq %r12, %rdi ...
/tpoisonooo[P]ncnn/src/layer/convolution1d.cpp
ncnn::Convolution1D::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Convolution1D::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { Mat bottom_blob_bordered; make_padding(bottom_blob, bottom_blob_bordered, opt); if (bottom_blob_bordered.empty()) return -100; const int w = bottom_blob_bordered.w; const size_t elemsize = bottom_bl...
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x48, %rsp movq %rcx, %r15 movq %rdx, %r14 movq %rdi, %rbx movq %rsp, %rdx andq $0x0, 0x40(%rdx) xorps %xmm0, %xmm0 movaps %xmm0, (%rdx) movups %xmm0, 0xc(%rdx) movaps %xmm0, 0x20(%rdx) movups %xmm0, 0x2c(%rdx) callq 0x1beb7e pushq $-0x64 popq %rbp cmpq $0x0, (%rsp) je ...
/tpoisonooo[P]ncnn/src/layer/convolution1d.cpp
ncnn::convolution1d(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Mat const&, int, int, int, int, ncnn::Mat const&, ncnn::Option const&)
static int convolution1d(const Mat& bottom_blob, Mat& top_blob, const Mat& weight_data, const Mat& bias_data, int kernel_w, int stride_w, int dilation_w, int activation_type, const Mat& activation_params, const Option& opt) { const int h = bottom_blob.h; const int outw = top_blob.w; const int outh = top_bl...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rdx, 0x40(%rsp) movl 0xd8(%rsp), %r10d movl 0xd0(%rsp), %eax movl 0x30(%rdi), %r11d movslq 0x2c(%rsi), %rbx movl 0x30(%rsi), %r14d movq (%rcx), %rdx movq %rdx, 0x58(%rsp) testq %rdx, %rdx je 0x1bebdd movslq 0x38(%rcx), %rdx imulq 0...
/tpoisonooo[P]ncnn/src/layer/convolution1d.cpp
ncnn::Convolution1D::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Convolution1D::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const Mat& _weight_data = bottom_blobs[1]; Mat& top_blob = top_blobs[0]; const int _kernel_w = _weight_data.w; const int _num_output = _...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xf8, %rsp movq %rcx, %r12 movq %rsi, %r15 movq %rdi, %rbx movq (%rsi), %rbp leaq 0x48(%rbp), %rdi movq (%rdx), %rax movq %rax, 0x8(%rsp) movl 0x74(%rbp), %r14d movl 0x80(%rbp), %eax movl %eax, 0x4(%rsp) leaq 0xb0(%rsp), %rsi andq $0x0, 0x40(%rsi) ...
/tpoisonooo[P]ncnn/src/layer/convolution1d.cpp
virtual thunk to ncnn::Convolution1D_x86::create_pipeline(ncnn::Option const&)
int Convolution1D_x86::create_pipeline(const Option& opt) { if (dynamic_weight) return 0; int num_input = weight_data_size / kernel_w / num_output; int elempack = 1; int out_elempack = 1; #if __SSE2__ if (opt.use_packing_layout) { #if __AVX__ elempack = num_input % 8 == 0 ? 8 ...
pushq %rax movq (%rdi), %rax addq -0x30(%rax), %rdi callq 0x1bf318 xorl %eax, %eax popq %rcx retq nop
/tpoisonooo[P]ncnn/src/layer/x86/convolution1d_x86.cpp
virtual thunk to ncnn::Convolution1D_x86::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Convolution1D_x86::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int w = bottom_blob.w; int h = bottom_blob.h; size_t elemsize = bottom_blob.elemsize; int elempack = bottom_blob.elempack; const int kernel_extent_w = dilation_w * (kernel_w - 1) + 1; Mat bottom_bl...
movq (%rdi), %rax addq -0x48(%rax), %rdi jmp 0x1bf5e2
/tpoisonooo[P]ncnn/src/layer/x86/convolution1d_x86.cpp
ncnn::ConvolutionDepthWise1D::forward(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int ConvolutionDepthWise1D::forward(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { Mat bottom_blob_bordered; make_padding(bottom_blob, bottom_blob_bordered, opt); if (bottom_blob_bordered.empty()) return -100; const int w = bottom_blob_bordered.w; const size_t elemsize = ...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x50, %rsp movq %rcx, %r15 movq %rdx, %r14 movq %rsi, %r12 movq %rdi, %rbx movq %rsp, %rdx andq $0x0, 0x40(%rdx) xorps %xmm0, %xmm0 movaps %xmm0, (%rdx) movups %xmm0, 0xc(%rdx) movaps %xmm0, 0x20(%rdx) movups %xmm0, 0x2c(%rdx) callq 0x1c8976 pushq $-0x64 popq...
/tpoisonooo[P]ncnn/src/layer/convolutiondepthwise1d.cpp
ncnn::get_cpu_thread_affinity_mask(int)
const CpuSet& get_cpu_thread_affinity_mask(int powersave) { setup_thread_affinity_masks(); if (powersave == 0) return g_thread_affinity_mask_all; if (powersave == 1) return g_thread_affinity_mask_little; if (powersave == 2) return g_thread_affinity_mask_big; NCNN_LOGE("po...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x138, %rsp # imm = 0x138 movl %edi, 0xc(%rsp) xorps %xmm0, %xmm0 movups %xmm0, 0x57806(%rip) # 0x2242c8 movups %xmm0, 0x577ef(%rip) # 0x2242b8 movups %xmm0, 0x577d8(%rip) # 0x2242a8 movups %xmm0, 0x577c1(%rip) # 0x224298 mov...
/tpoisonooo[P]ncnn/src/cpu.cpp
Coverage_mathCnWithNamespaceViolations_Test::TestBody()
TEST(Coverage, parserBranchesCMetaIdCellml2) { libcellml::ParserPtr parser = libcellml::Parser::create(); auto model = parser->parseModel(fileContents("coverage/cmeta_id_cellml2.xml")); EXPECT_EQ(size_t(0), model->unitsCount()); EXPECT_EQ(size_t(0), model->componentCount()); }
pushq %r15 pushq %r14 pushq %rbx subq $0x60, %rsp leaq 0x40(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0x153b7(%rip), %rsi # 0x25a3e leaq 0x155f3(%rip), %rdx # 0x25c81 leaq 0x30(%rsp), %rbx movq %rbx, %rdi callq 0x21b06 leaq 0x20(%rsp), %r14 movq %r14, %rdi movl $0x1, %esi callq 0xdd10 movq (%r14), %rsi leaq 0x50(...
/cellml[P]libcellml/tests/coverage/coverage.cpp